Scholarlink

<p>Flood susceptibility mapping using geoInformation and machine learning-based models is of vital importance to predict future flood occurrences and make informed decisions on mitigation strategies. This study aims to assess the applicability of three widely used machine learning models, Classification and Regression Trees (CART), Support Vector Machines (SVM), and Extreme Gradient Boosting (XGBoost), and to evaluate their performance in mapping flood susceptibility in the Tensift Watershed, located in the central-western part of Morocco within the Marrakech province. Sixteen conditioning factors spanning topographic, geologic, climatic, and land cover domains were used as model inputs. A total of 228 flood inventory points, consisting of 114 flood and 114 non-flood locations, were used to train and test the models. The area under the receiver operating characteristic curve (AUC) was used to assess the performance of models. The results indicate that the CART model achieved the highest predictive accuracy with an AUC of 0.882, followed by SVM (AUC = 0.860) and XGBoost (AUC = 0.833). These findings suggest that CART is the most suitable model for flood susceptibility assessment in the study area. The outcomes of this research are also expected to support flood risk management activities in Marrakech province by identifying vulnerable areas, guiding infrastructure planning, and enhancing community preparedness to minimize adverse impacts of future flood events on both people and the environment.</p> Graphical Abstract <p>This graphical abstract provides a visual summary of the methodology and key findings of a machine learning–based flood susceptibility mapping in one of the most flood-prone areas in central Morocco, the Tensift watershed (left panel). Along with intensive flood inventory mapping, high-resolution datasets obtained from diverse remote sensing products were used to extract 16 topographic, hydroclimatic, geologic, and anthropogenic conditioning factors, including slope, elevation, plan curvature, aspect, distance to roads, distance to faults, distance to streams, rainfall, normalized difference vegetation index, land cover, lithology, soil, sediment topographic wetness index, sediment transport index, stream power index, terrain ruggedness index, and transport index (central-bottom panel). The study employed three machine learning models: Classification and Regression Trees (CART), Extreme Gradient Boosting (XGBoost), and Support Vector Machine (SVM) (central-top panel). Accuracy assessment results showed that CART was the best-performing model with an AUC of 88.2%, followed by SVM and XGBoost, each with an average AUC of 83.3% (central-right panel). Similarly, CART predicted the highest percentage area for the high (20.61%) and very high (20.97%) susceptibility classes, while predicting the lowest percentage area for the very low (20.02%), low (20.16%), and moderate (18.24%) classes compared to the other models (top-right panel). The findings underscore the applicability of machine learning algorithms for mapping flood susceptibility to support future land and water resource management in Morocco.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine Learning-Based Flood Susceptibility Mapping Using Geoenvironmental Factors in Central Morocco

  • 期刊论文

  • 发表日期：

    2026年2月1日

查看全文
  • Meriame Mohajane,
  • Sk Ajim Ali,
  • Sliman Hitouri,
  • Renata Pacheco Quevedo,
  • Tadesual Asamin Setargie,
  • Costanza Fiorentino,
  • Ismail ElKhrachy,
  • Paola D’Antonio,
  • Meriam Lahsaini

摘要

Flood susceptibility mapping using geoInformation and machine learning-based models is of vital importance to predict future flood occurrences and make informed decisions on mitigation strategies. This study aims to assess the applicability of three widely used machine learning models, Classification and Regression Trees (CART), Support Vector Machines (SVM), and Extreme Gradient Boosting (XGBoost), and to evaluate their performance in mapping flood susceptibility in the Tensift Watershed, located in the central-western part of Morocco within the Marrakech province. Sixteen conditioning factors spanning topographic, geologic, climatic, and land cover domains were used as model inputs. A total of 228 flood inventory points, consisting of 114 flood and 114 non-flood locations, were used to train and test the models. The area under the receiver operating characteristic curve (AUC) was used to assess the performance of models. The results indicate that the CART model achieved the highest predictive accuracy with an AUC of 0.882, followed by SVM (AUC = 0.860) and XGBoost (AUC = 0.833). These findings suggest that CART is the most suitable model for flood susceptibility assessment in the study area. The outcomes of this research are also expected to support flood risk management activities in Marrakech province by identifying vulnerable areas, guiding infrastructure planning, and enhancing community preparedness to minimize adverse impacts of future flood events on both people and the environment.

Graphical Abstract

This graphical abstract provides a visual summary of the methodology and key findings of a machine learning–based flood susceptibility mapping in one of the most flood-prone areas in central Morocco, the Tensift watershed (left panel). Along with intensive flood inventory mapping, high-resolution datasets obtained from diverse remote sensing products were used to extract 16 topographic, hydroclimatic, geologic, and anthropogenic conditioning factors, including slope, elevation, plan curvature, aspect, distance to roads, distance to faults, distance to streams, rainfall, normalized difference vegetation index, land cover, lithology, soil, sediment topographic wetness index, sediment transport index, stream power index, terrain ruggedness index, and transport index (central-bottom panel). The study employed three machine learning models: Classification and Regression Trees (CART), Extreme Gradient Boosting (XGBoost), and Support Vector Machine (SVM) (central-top panel). Accuracy assessment results showed that CART was the best-performing model with an AUC of 88.2%, followed by SVM and XGBoost, each with an average AUC of 83.3% (central-right panel). Similarly, CART predicted the highest percentage area for the high (20.61%) and very high (20.97%) susceptibility classes, while predicting the lowest percentage area for the very low (20.02%), low (20.16%), and moderate (18.24%) classes compared to the other models (top-right panel). The findings underscore the applicability of machine learning algorithms for mapping flood susceptibility to support future land and water resource management in Morocco.